Access to AI project management tools has grown 50% year on year, according to McKinsey’s Superagency in the Workplace report. The outcomes haven’t kept pace. Only 1% of companies describe themselves as mature in AI deployment. Among US C-suite respondents, just 19% reported revenues had increased by more than 5% from AI. On costs, only 23% reported any favourable movement.
That gap matters most in project and task management, where every major platform spent 2025 and early 2026 repositioning around agents. The pitch is consistent: less manual coordination, fewer status updates written by hand, decisions surfaced before they become delays. The reality, for most IT buyers, is more complicated.
What the Platforms Are Actually Shipping
Monday.com repositioned its entire platform around native AI agents in May 2026, rebuilding its permissions model and data layer on the assumption that agents will do real work, not just assist humans doing it. Its AI meeting assistant joins calls and creates action items directly in monday boards. One-click connectors link to Microsoft 365 Copilot, OpenAI’s ChatGPT, and Google Gemini. Asana has introduced AI Teammates for agentic collaboration, built on its Work Graph so agents understand project dependencies from the start. ClickUp offers Super Agents that watch for triggers and execute multi-step workflows without human input. Adobe Workfront now lets project managers add AI agents to plans as assignable resources, the same as a human team member.
Microsoft has meanwhile used Planner’s AI updates and the retirement of Project Online to consolidate its work management offering, with Copilot licensing as the gateway to more advanced automation. Monday.com, Asana, and Smartsheet are all actively targeting the migration window this creates.
Deloitte’s 2026 State of AI in the Enterprise report, based on 3,235 senior leaders, found that only 25% of organisations have moved 40% or more of their AI pilots into production, and just 34% report using AI to deeply transform their business. The remaining third are using it at surface level, with little change to existing processes.
The Problem Is Not the Software
The most useful detail in Deloitte’s research is where the failures are concentrated. It is not model quality or feature gaps. It is the same infrastructure and data problem that surfaces in every other category of enterprise AI deployment. As we covered in the context of ITSM, agents need clean, connected data to act reliably. In project management, that means tasks, dependencies, ownership, and status information that is consistently structured and up to date. Most enterprise project estates are not.
The practical advice buried in monday.com’s own release documentation makes the point plainly: AI features are most effective when the underlying data is clean and consistently structured. If boards have inconsistent column naming, scattered ownership, or outdated statuses, AI agents will surface that chaos rather than resolve it. Asana’s documentation says something similar about AI Studio: teams that skip mapping their current process before building an AI workflow often build automation around a broken process and are then confused when the output is inconsistent.
Gartner projects that over 40% of agentic AI projects will be at risk of cancellation by 2027 if proper governance controls are not in place. IDC adds that companies failing to establish AI-ready data foundations before scaling will see a 15% productivity loss by 2027.
What IT Buyers Should Be Evaluating
For IT leaders assessing AI project management tools, the vendor AI roadmap is the wrong place to start. The more useful questions are operational.
Is project data consistently structured across the organisation? Agents summarise and act on the data they can see. Inconsistent tagging, stale status fields, and tasks created outside the primary tool all degrade output quality before the agent has made a single decision.
How does each platform handle AI governance? Asana logs and audits every AI action. Monday.com operates on a credit model that can pause AI features under heavy usage. ClickUp’s Super Agents offer more automation depth but require more configuration to operate safely at scale. These are not minor product differences. For enterprise IT teams responsible for audit trails and compliance, they are procurement criteria.
What is the integration model with the existing UC stack? Microsoft’s April 2026 Copilot Studio updates brought monday.com, Asana, and ServiceNow into Copilot Chat as native agent experiences, meaning enterprise teams already running M365 can surface project data and take action inside Teams conversations. For organisations not on M365, that integration advantage largely disappears, and the choice of project management tool becomes more dependent on native API coverage and third-party connector depth.
The underlying issue is one the market has been slow to name directly. Buying a platform with an AI agent does not make project data AI-ready. The two things are separate work, and the second one has to happen first.
